init-architect
Pass
Audited by Gen Agent Trust Hub on Feb 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (LOW): Vulnerable to Indirect Prompt Injection (Category 8). Ingestion points: Analyzes repository structure and existing architecture markdown files. Boundary markers: No delimiters or safety instructions provided for external data processing. Capability inventory: Can create and update .json and .md files in the repository root and skill directory. Sanitization: No validation or sanitization of input data from the codebase.
- [DATA_EXPOSURE] (SAFE): Reads repository metadata and identifies sensitive architectural areas for documentation; no data exfiltration detected.
- [COMMAND_EXECUTION] (SAFE): Performs local file system operations (read/write) consistent with documentation and configuration management.
Audit Metadata